FEW-SHOT LEARNING: REVIEW OF APPLICATIONS IN HEALTH AND AGRICULTURE

Öz

Few-Shot Learning (FSL) is an innovative machine learning method that aims to achieve high accuracy with limited data. In this study, we focus on the applications of FSL in healthcare and agriculture, where data collection is difficult and costly. In the health sector, FSL provides high accuracy with little data in skin cancer diagnosis and eye disease diagnosis. In the agriculture sector, it is used in the early diagnosis of plant diseases to increase crop productivity. The study examines FSL's meta-learning, transfer learning and similarity-based techniques to assess its advantages over data constraints and its potential for future applications.

Anahtar Kelimeler

• Few-Shot Learning
• Meta-learning
• Transfer learning
• Similarity-based learning
• Data limitation
• Medical imaging
• Plant disease diagnosis
• Skin cancer diagnosis
• Agriculture practices

AZ ATIŞLI ÖĞRENME: SAĞLIK VE TARIM ALANINDA UYGULAMALARIN İNCELENMESİ

Öz

Few-Shot Learning (FSL), sınırlı veriyle yüksek doğruluk elde etmeyi hedefleyen yenilikçi bir makine öğrenimi yöntemidir. Bu çalışmada, FSL'nin veri toplamanın zor ve maliyetli olduğu sağlık ve tarım alanındaki uygulamaları ele alınmaktadır. Sağlık sektöründe cilt kanseri teşhisi ve göz hastalıklarının tanısında FSL, az veri ile yüksek doğruluk sağlamaktadır. Tarım sektöründe ise bitki hastalıklarının erken teşhisinde kullanılarak ürün verimliliğini artırmaktadır. Çalışma, FSL’nin meta-öğrenme, aktarım öğrenmesi ve benzerlik tabanlı tekniklerini inceleyerek veri kısıtlılığına karşı sunduğu avantajları ve gelecekteki uygulama potansiyelini değerlendirmektedir.

Anahtar Kelimeler

• Few-Shot Learning
• Meta-öğrenme
• Aktarım öğrenmesi
• Benzerlik tabanlı öğrenme
• Veri kısıtlılığı
• Tıbbi görüntüleme
• Bitki hastalığı tanısı
• Cilt kanseri teşhisi
• Tarım uygulamaları
• Sağlık uygulamaları

Kaynakça

1. Argüeso, D., Picon, A., Irusta, U., Medela, A., San-Emeterio, M. G., Bereciartua, A. ve Alvarez-Gila, A. (2020). Few-shot learning approach for plant disease classification using images taken in the field. Computers and Electronics in Agriculture, 175: 105542.
2. Aybar, M. Talaş, U. ve Çubukçu, B. (2024). Transfer öğrenme modelleri ile elma yapraklarında hastalık tespiti. ESTUDAM Bilişim, 5(2): 57–63. doi: 10.53608/estudambilisim.1556425.
3. Baik, S., Yu, H. ve Rusu, A. A. (2021). Meta-learning with task-adaptive loss function for few-shot learning. International Conference on Computer Vision (ICCV), 8500-8509.
4. Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., ... ve Amodei, D. (2020). Language models are few-shot learners. Advances İn Neural İnformation Processing Systems, 33: 1877-1901.
5. Camargo, A. ve Smith, J. S. (2009). An image-processing based algorithm to automatically identify plant disease visual symptoms. Biosystems Engineering, 102(1): 9-21.
6. Chao, S. ve Belanger, D. (2024). Generalizing whole-genome doubling classification. Harvard University.
7. Cheng, H., Wang, Y., Li, H., Kot, A. C. ve Wen, B. (2023). Disentangled feature representation for few-shot image classification. IEEE Transactions On Neural Networks And Learning Systems.
8. Elsken, T., Staffler, B., Metzen, J. H. ve Hutter, F. (2020). Meta-learning of neural architectures for few-shot learning. CVPR.

9. Ergün, E. ve Kılıç, K. (2021). Derin öğrenme ile artırılmış görüntü seti üzerinden cilt kanseri tespiti. Black Sea Journal of Engineering and Science, 4(4): 192-200. https://doi.org/10.34248/bsengineering.938520
10. Esen, F. A. ve Onan, A. (2022). Derin öğrenme yöntemleri ile bitki yaprakları üzerindeki hastalıkların sınıflandırılması. 40: 151-155.
11. Ge, Y., Guo, Y., Yang, Y., vd. (2024). Few-shot learning for medical text: A systematic review. Emory University.
12. Gidaris, S. ve Komodakis, N. (2019). Boosting few-shot visual learning with self-supervision. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 8040-8050.
13. Hosny, K. M., Kassem, M. A. ve Foaud, M. M. (2018). Skin cancer classification using deep learning and transfer learning: 9th Cairo International Biomedical Engineering Conference (CIBEC) (pp. 90-93). Cairo. https://doi.org/10.1109/CIBEC.2018.8641762
14. Işık, G. ve Paçal, İ. (2024). Meta-learning for breast cancer ultrasound classification. Neural Computing.
15. Kaya, V. ve Akgül, İ. (2023). Classification of skin cancer using VGGNet model structures. Gümüşhane Üniversitesi Fen Bilimleri Dergisi, 13(1): 190-198.
16. Khamparia, A. ve Pandey, B. (2021). Hybrid deep learning model for predicting plant diseases using image processing techniques. Journal of Ambient Intelligence and Humanized Computing, 12(2): 2023-2032. https://doi.org/10.1007/s12652-020-02366-8
17. Lake, B. M., Salakhutdinov, R. ve Tenenbaum, J. B. (2015). Human-level concept learning through probabilistic program induction. Science.
18. Li, K., Zhang, Y., Li, K. ve Fu, Y. (2020). Adversarial feature hallucination networks for few-shot learning. CVPR.
19. Li, Y. ve Chao, X. (2021). Semi-supervised few-shot learning approach for plant diseases recognition. Plant Methods, 17: 68. https://doi.org/10.1186/s13007-021-00770-1
20. Li, Y. ve Yang, J. (2020). Few-shot cotton pest recognition and terminal realization. Computers and Electronics in Agriculture, 169: 105240. https://doi.org/10.1016/j.compag.2020.105240
21. Lin, C., Wei, D., Cheng, Y., vd. (2024). Few-shot learning for lung cancer metastasis classification. NYCU.
22. Liu, H., Wang, C., Jiang, X. ve Khishe, M. (2024). A few-shot learning approach for COVID-19 diagnosis. JAISCR.
23. Mahajan, K., Sharma, M. ve Vig, L. (2024). Meta-DermDiagnosis for skin disease identification. CVPRW.
24. Munkhdalai, T. ve Yu, H. (2017). Meta Networks. International Conference on Machine Learning (ICML) içinde 70: 2554-2563.
25. Ngugi, L. C., Abelwahab, M. ve Abo-Zahhad, M. (2021). Recent advances in image processing techniques for automated leaf pest and disease recognition – A review. Information Processing in Agriculture, 8: 27–51. https://doi.org/10.1016/j.inpa.2020.04.004
26. Parnami, A. ve Lee, M. (2022). Learning from few examples: A summary of approaches to few-shot learning. arXiv preprint arXiv:2203.04291. https://arxiv.org/abs/2203.04291
27. Protonotarios, N., Katsamenis, I., Sykiotis, S., vd. (2024). A few-shot U-Net deep learning model. Biomedical Physics.
28. Qiao, L., Shi, Y., Li, J., Wang, Y., Huang, T. ve Tian, Y. (2019). Transductive episodic-wise adaptive metric for few-shot learning. International Conference on Computer Vision (ICCV). Retrieved from [source](9Laplacian Regularized …).
29. Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., ... ve Liu, P. J. (2020). Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(140): 1-67.
30. Ravi, S. ve Larochelle, H. (2017). Optimization as a model for few-shot learning. International Conference On Learning Representations (ICLR).
31. Ren, M., Triantafillou, E., Ravi, S., Snell, J., Swersky, K., Tenenbaum, J. B., Larochelle, H. ve Zemel, R. S. (2018). Meta-learning for semi-supervised few-shot classification. International Conference on Learning Representations (ICLR).
32. Ristaino, J. B., Anderson, P. K., Bebber, D. P., Brauman, K. A., Cunniffe, N. J., vd. (2021). The persistent threat of emerging plant disease pandemics to global food security. Proceedings of the National Academy of Sciences, 118(23): e2022239118. https://doi.org/10.1073/pnas.2022239118
33. Santoro, A., Bartunov, S., Botvinick, M., Wierstra, D. ve Lillicrap, T. (2016). Meta-learning with memory-augmented neural networks. International Conference on Machine Learning (ICML), 1842-1850.
34. Schick, T. ve Schütze, H. (2021). It’s not just size that matters: Small language models are also few-shot learners. arXiv preprint arXiv:2009.07118.
35. Schwartz, E., Karlinsky, L., vd. (2018). Delta-encoder: An effective sample synthesis method. NeurIPS.
36. Snell, J., Swersky, K. ve Zemel, R. (2017). Prototypical networks for few-shot learning. Advances in Neural Information Processing Systems (NeurIPS), 30: 4077-4087.
37. Sun, Q., Liu, Y., Chua, T. S. ve Schiele, B. (2018). Meta-transfer learning for few-shot learning. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 403-412.
38. Sung, F., Yang, Y., vd. (2018). Relation networks for few-shot learning. CVPR.
39. Upadhyay, N., Bhargava, A., Singh, U., vd. (2024). Enhancing breast cancer classification with DenseNet-121. medRxiv.
40. Vinyals, O., Blundell, C., Lillicrap, T., Kavukcuoglu, K. ve Wierstra, D. (2016). Matching networks for one shot learning. Advances in Neural Information Processing Systems (NeurIPS), 3630-3638.
41. Wang, M., Liu, W., Gu, X., vd. (2024). Few-shot learning for endometrial cancer detection. Heliyon.
42. Yang, J., Guo, X., Li, Y., Marinello, F., Ercisli, S., ve Zhang, Z. (2022). A survey of few-shot learning in smart agriculture: developments, applications, and challenges. Plant Methods, 18(28). https://doi.org/10.1186/s13007-022-00866-2
43. Zhang, C., Cai, Y., Lin, G. ve Shen, C. (2020). DeepEMD for few-shot ımage classification. CVPR.
44. Ziko, I. M., Boudiaf, M., Dolz, J., Granger, E. ve Ayed, I. B. (2021). Transductive few-shot learning: Clustering is all you need?. arXiv preprint arXiv:2106.09516